Why Multi-Chain Quadratic Funding Incentivizes Sybil Attacks

Sybil attacks are profitable in multi-chain QF because the funding formula squares the sum of contributions. A single entity splitting capital into many fake identities across chains like Arbitrum and Optimism amplifies its matching weight exponentially, not linearly.

Cross-chain identity is fractured. Current solutions like Gitcoin Passport or BrightID operate per-chain or off-chain, failing to provide a global, sybil-resistant identity layer. An attacker's reputation on Ethereum is irrelevant on Avalanche.

The matching pool is a target. Protocols like Optimism's RetroPGF allocate millions from a communal treasury. Sybil farmers treat this pool as a yield source, deploying bots to farm grants across every supported L2 and alt-L1.

Evidence: The 2023 Gitcoin Grants Round 18 saw a 33% sybil rate. In a multi-chain future with fragmented identity, this rate will approach 100% without new infrastructure.

Sybil attack costs plummet when identity verification is siloed per chain. An attacker creates a single identity on Ethereum, then replicates it cheaply on Arbitrum, Optimism, and Polygon. The cost of a fraudulent identity is the gas fee on the cheapest chain, not the cumulative cost across all chains.

Matching fund leakage accelerates because the QF algorithm cannot distinguish between a unique human on one chain and its Sybil clones on ten others. Each clone's small, cheap donation triggers matching funds from a shared pool, draining capital with minimal capital outlay.

Cross-chain attestation is insufficient. Tools like Gitcoin Passport or World ID rely on centralized oracles or bridges like LayerZero for state synchronization, creating new trust assumptions and latency that break QF's real-time, trust-minimized requirements.

Evidence: A 2023 simulation by EigenLayer researchers showed that a $1,000 investment in Sybil identities across 5 chains could extract over $50,000 from a naive multi-chain QF pool, a 50x ROI that makes fraud inevitable.

The subsidy is the exploit. Quadratic funding (QF) matches donations based on the square root of unique contributors, not total capital. This creates a matching subsidy per identity. On a single chain, creating a new Sybil identity costs gas. On multiple chains, a user with funds on Arbitrum and Optimism is two unique, subsidizable identities for the cost of one bridge transaction via Across or Stargate.

Cross-chain fragmentation breaks the cost barrier. The original QF model assumed a high identity creation cost (e.g., Ethereum mainnet gas). Multi-chain ecosystems like Polygon, Avalanche, and Base lower this cost to near-zero. An attacker's capital on one chain funds Sybil donations on ten others, exploiting ten separate matching pools for the price of bridge latency and fees.

The math guarantees positive ROI. The attack is not speculative. If the matching subsidy for a $1 donation exceeds the bridging and transaction cost, the attack prints risk-free money. Protocols like Gitcoin Grants see this empirically: Sybil clusters consistently arbitrage the difference between the subsidy value and the multi-chain operational cost, draining matching funds from legitimate projects.

Sybil attacks are economically rational when the cost to create a fake identity is lower than the expected reward. Quadratic funding amplifies small contributions, making this reward substantial. Gitcoin Passport stamps are a cost, not a barrier.

Attestations are not sybil-proof. Services like Ethereum Attestation Service (EAS) or Verax verify attributes (e.g., 'has a Twitter account'), not uniqueness. A sophisticated attacker can cheaply farm these credentials across multiple chains.

The multi-chain landscape fragments defense. A Passport score built on Optimism provides no sybil resistance for a funding round on Base or Scroll. Attackers exploit the weakest identity link across the ecosystem.

Evidence: Gitcoin Grants rounds, even with Passport, require complex, manual sybil filtering post-round. This proves the automated, on-chain solution does not exist. The cost to bypass a 'humanity score' is often less than $50.